Local DNS Asking for Remote Pass

Server Version#: Version 4.146.0
Player Version#: Browser

I’m running the latest Plex in Docker with --network=host on a local server (10.0.0.2). When I access it from my laptop 10.0.0.69 directly via 10.0.0.2:32400, it works perfectly and is recognized as a local/LAN connection.

But when I use a local DNS name set up through Pi-hole (e.g., plex.something.lan pointing to 10.0.0.2), Plex treats the connection as remote access and asks for the remote pass— even though it’s on the same network.

Is there any way to make Plex treat this as a local connection while still using the DNS name?

Thanks!

I don’t think so.

To add to Tom,

If your local DNS server only resolves Hostname → IP without expanding hostname to FQDN then you’re ok.

If you’re using FQDN in the URL, the browser will use that FQDN as it connects.

The problem stems from PMS seeing the FQDN in the “Referrer” fields.
Whenever PMS sees a FQDN in these fields, whether local or not, it treats it as Remote.

e.g

  • HostnameLAN.IP.of.Server = :white_check_mark:
  • HostnameFQDN → `LAN.IP.of.Server’ = :x:
  • FQDNLAN.IP.of.Server = :x:

No matter what, even hostname directly to IP is considered remote access. E.g. arch -> 10.0.0.69.

I saw that. How isn’t this considered a BUG?

The whole log shows it as a local connection, but here:

Jul 04, 2025 15:13:44.162 [140204296354616] Debug — Completed: [10.0.0.69:42238] 302 GET /web (8 live) #cd74 GZIP 0ms 239 bytes (pipelined: 1) -> http://somehostname:32400/web/index.html

Jul 04, 2025 15:13:45.192 [140204177730360] Debug — WebSocket: Performing handshake from origin [http://somehostname:32400](http://somehostname:32400/)

Everything else comes from 10.0.0.69:XXXXX (Subnet).

It seems like a really questionable design choice to treat this as remote access.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.